Putz Z in lmer

From: Douglas Bates <bates at stat.wisc.edu>
To: Eder David Borges da Silva <eder at leg.ufpr.br>
Cc: R-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] Putz Z in lmer
Message-ID:
? ? ? ?<CAO7JsnRRhpvQ5JyQSVHHw-=WCptwRxeSST5OeudC9S_bFhu+rQ at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1

On Wed, Aug 17, 2011 at 3:09 PM, Eder David Borges da Silva
<eder at leg.ufpr.br> wrote:

Dear group members
Imagine a mixed model y = Xb + Zu + e, Xb = fixed, random = Zu, is possible
in lme4: lmer declare the matrix of random effects for the adjustment??
Is there a function that allows to adjust this model stating the matrix Z and
X?

Not directly and, even if you could, it would not be sufficient to
specify Z - you also need to specify the form of the
variance-covariance matrix of the random effects and how it is
generated from the parameters to be optimized.

We do plan to provide facilities to do this in a future release.



------------------------------